- Natural Language Processing Class with LingPipe--Care and Feeding of Classifiers
Classifiers take some sort of input, text for us, and assign it to one or more categories. This very simple capability forms the core of modern NLP techniques like sentiment analysis, named entity detection and information extraction. This class will cover both heuristic and machine learning based classifiers from the perspective of a programmer building such systems. LingPipe is an open source API designed to build NLP systems. Breck Baldwin instructs the class. The class covers rarely addressed issues around creation, training and maintaining classifiers with a focus on modern learning techniques like logistic regression. This is a practitioner level course with minimal theory appropriate for proficient developers. The class will be taught with real time coding and those that wish can follow along on their own laptop. The course outline is: - Introduction to classifiers and information they provide ranging from first best classifiers to conditional probabilities. - Creation of gold standard data with tweets - Classifier evaluation a.k.a. the dreaded evaluation harness - Heuristic classifiers - Machine learning classifiers - Feature tuning for machine learning - Common problems encoded into a classifier framework The class is 6:30pm to 9:30pm Tuesday, October 25 at 181 N 11th St, Suite 401, Brooklyn NY 11211. The location is 4 blocks from the Bedford L stop and is a 4 floor walkup. There will be pizza served as a break and those that wish to have a beer afterwards will meet at the famed Mugs Ale House a block away. The cost is $100 with a class limit of 15. Registration closes October 23. Payment via PayPal to [masked].
- The Poster Session: Semantic Web and Knowledge Engineering Social Night
For the latest information, go here: http://thepostersession.github.io/ The Semantic Web and Knowledge Engineering meetup social night with posters and demos from area companies. The purpose of this event is primarily social, to bring together people in the NYC area who are interested in these subjects, and provide a low-key forum to let everyone know what you’re doing. A “poster” is simply a 3’x4’ board on which you have information about what your company does. During the 2 hour session, many people will choose to stay close to their poster much of the time, in order to answer questions. Presenting a poster is not required to attend, but presenting is free and gives you an excuse to say what you do, so why not do it! The meeting space is limited, we will prefer poster presenters to regular attendees if we get more than 150 registrants. To register as an attendee only: Lotico New York Semantic Web http://www.meetup.com/semweb-25/ New York Knowledge Engineering http://www.meetup.com/New-York-Knowledge-Engineering-Meetup/ To register as a Presenter, please follow these steps: Create an account at https://www.easychair.org Once you are logged-in, please go to the PS2015 page: https://easychair.org/conferences/?conf=ps2015 Click on “New Submission” and complete the form. Please be as clear in your abstract as possible so we have an understanding of what your poster is about. While we expect some sales orientation, the objective is to share expertise and applications of these semantic web and knowledge engineering technologies. Poster submissions will be reviewed only for relevance and pitch - please avoid sales presentations and instead focus on communicating what you do and why semantic web and knowledge engineering meetup members would be interested. Demos are encouraged but not required. If you plan to give a demo please let us know in your submission so we can plan for it. Due to space limitations, we may have to limit acceptance. Deadline for presenters is November 9, 2015 (1 week prior to the event). You can submit as many poster proposals as you like, and have up to 3 authors at your discretion. All authors will be registered for the event. If you want others to attend, they may RSVP via these Meetup pages: Lotico New York Semantic Web http://www.meetup.com/semweb-25/ New York Knowledge Engineering http://www.meetup.com/New-York-Knowledge-Engineering-Meetup/ You will be provided with space to place your Poster on a table, chair or an easel. Please make it from cardboard or other sturdy material; posters will not be taped, pinned, or hung. We suggest you follow some of the tips on “how to make great poster presentations” using some of these links: https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#q=how%20to%20make%20great%20poster%20presentations You may also desire to use a laptop or tablet for your presentation or demonstration. Large Audio Visual equipment (projectors, TV, monitors) will not be permitted to be brought into this event, sorry. Free WiFi and electricity will be available. DETAILS Date: Monday, November 16th, 2015 Time: Doors open at 6:00pm. Event starts at 6:30pm and ends at 8:30pm RSVP: is required for this free event. You must register with your real name and bring ID to enter the building Location: Google, 111 Eighth Avenue, Fourth Floor, New York, New York 10011. Please enter from the north side, nearest 16th Street on 8th Ave. The event is free of charge, with food and adult drinks provided. Deadlines Poster, demo submissions and RSVPs close November 9th. Posters and demos can be submitted here: https://easychair.org/account/signin.cgi?key=29962407.pMUcAi98PP0U1heB
- ISWC-2015 Poster&Demo Reception
The ISWC organizers have agreed to open up the posters & demo reception evening to NY Semantic Web and Knowledge Engineering Meetup members. The posters & demo night is really the best and most exciting session during the conference, and the easiest way to get the pulse of the field. Please see the ISWC 2015 website for more information on the venue and how to get there. Registration will be $50 and includes light food and some free drinks. If you plan to attend, you should consider registering for the full conference.
- Francois Scharffe on Interlinking items, categories and linked datasets
Title: Interlinking items, categories and linked datasets Room: 109 Abstract: 3Top is a ranking and recommendation platform based on a simple model: users choose or enter a category, and one to three items representing their top choices for this category. "3Top sushi restaurants in the West Village" or "3Top science fiction movies set on the moon" are such categories. Our system processes the categories and items to provide a better user experience: additional information and media for items, links to third-party rating websites and vendors, and category and item recommendations based on a user’s prior rankings. In this presentation, I will show how we use semantic web technologies combined with natural language processing to build an intelligent platform making sense of categories and items. I will present the modular ontology serving as the pivot between various linked datasets, and the techniques we are developing to link from natural language input to ontology concepts and web resources. Speaker Bio: After more than 10 years of academic research on semantic web and linked open data, ontology alignment and data interlinking, Francois Scharffe is now co-founder and knowledge engineer at 3Top. http://www.3top.com (http://www.3top.com/)
- Pat Hayes and Margaret Warren discuss Image Snippets
Join us for another interesting discussion from Pat Hayes, a progenator of both AI and the Semantic Web, and Margaret Warren, an artist and entrepreneur. The meetup will be hosted by Google at no cost. Enter on 8th Ave. near 15th St (111 8th Ave). The meeting will be on the 5th Floor "Water Tower Cafe". Abstract Deploying semantic web standards and technologies to image annotation is a challenge in both theory and practice. Formalisms based on RDF triples can be useful in image search, particularly if the structured information about image content can be captured from the annotator at the information recording point. ImageSnippets is a general purpose tool we have developed that allows untrained users to apply semantic annotation to images. Users are not required to have a deep knowledge of the rather arcane syntax of RDF or OWL. The design of this system has led us to consider many semantic web issues which we will discuss including the relative merits and uses of ‘deep’ versus ‘shallow’ ontology content; how to find concept identifiers on the Web (and what to do when you don’t find them); how to combine domain expertise with image annotation skills; and the importance of deep knowledge in ‘niche’ image curation. Patrick Hayes Pat has a long research career in Artificial Intelligence, with an emphasis on logical knowledge representation. Significant early milestones include one of the first papers describing what is now called logic programming and the first paper describing what is now called ontology design. More recently, Pat was a member of the W3C Working groups which defined the semantic web standards RDF, OWL and SPARQL, and co-authored the RDF and OWL semantics specifications. Pat has been President of AAAI, Chairman of IJCAI and Associate Editor of the AI Journal, and has held university Chairs in Computer Science, Philosophy and Cognitive Science. Pat is the author of numerous technical and popular publications and presentations at more conferences and meetings than anyone can now count. He is also a visual artist with work in several galleries and collections. Margaret Warren Margaret has had a diverse career in the computing sciences, business management and art. While she has always been an artist/photographer, Margaret worked with cryptographic electronics in the US Coast Guard and several early multimedia technology companies in programming, networking and systems administration. She operated her own IT support company for over 17 years and is currently acting as general manager for a vintage Porsche restoration company. But she has always been passionate about image description and has worked both personally and professionally with numerous annotation systems throughout her life and career. In an effort to improve these systems, she began researching the application of knowledge representation theory to image annotation and this work has now led to several academic publications and a system called ImageSnippets.
- From hyperlinks to semantic relations using Open Knowledge Extraction.
Come join us for some social interaction a a featured talk by Valentina Presutti and Aldo Gangemi at the 120 Park Ave. Bloomberg office. !!!CHANGE OF VENUE!!!! Please come to: Event Space - 1st Floor at 353WestNYC353 West 46th Street, Between 8th & 9th Ave, New York, NY 10036, US !!!CHANGE OF VENUE!!! Abstract: The vision of the Semantic Web is to populate the web with machine understandable information so that artificial intelligences (AI) can use it as background knowledge for assisting humans in performing a significant number of their daily tasks. Research in this field produced a standardised knowledge representation format (namely, linked data) and huge amount of machine-readable data available on the web (namely, the web of data), mostly derived from structured data (typically databases) or semi-structured data (e.g. Wikipedia infoboxes). However, most of the web consists of natural language text containing valuable knowledge for enriching the web of data. Hence, a main challenge is to extract as much relevant knowledge as possible from this content, and publish them in the form of linked data. Open Information Extraction (OIE) has been developed recently as an approach to extract information from unstructured data, mostly of a textual nature However, the information extracted is typically in the form of triples of strings (subject, relational phrase, object). OIE approaches are useful but insufficient alone for populating the web with machine readable information as their results are not directly linkable to, and immediately reusable from, other linked data sources. In this seminar, after giving a brief introduction to background concepts and notions, I will describe a work that proposes a novel Open Knowledge Extraction approach that performs unsupervised, open domain, and abstractive knowledge extraction from text for producing directly usable machine readable information. In particular I will discuss an approach based on the hypothesis that hyperlinks (either created by humans or knowledge extraction tools) provide a pragmatic trace of such semantic relations between two entities, and that such semantic relations, their subjects and objects, can be revealed by processing their linguistic traces (i.e. the sentences that embed the hyperlinks) and formalised as linked data and ontology axioms. Experimental evaluations conducted with the help of crowdsourcing confirm this hypothesis showing very high performances. A demo of Open Knowledge Extraction at http://wit.istc.cnr.it/stlab-tools/legalo.
- Cognitive Computing Christmas
Join the NYC semantic web meetup at WeWork on Fulton & Broadway to hear about Cognitive Computing, Crowdsourcing, and enabling the future of media. Plus holiday festivities... Chris Welty, Research Scientist at Google Research in NY Silver Oliver, Information Architect at Datalanguage Todd Carter, Founder/CEO at Tagasauris, Inc. Seating is limited, you must register in advance. Registration is $10 to cover the meeting space, food, beer & wine. For more information please RSVP. Thanks and we look forward to seeing you there.
- Semantic Technologies In Financial Services (STIFS)
- Semantic Technology in Publishing and Media
Join us for a set of short talks and a discussion on applications of Semantic Technologies in leading publishing and media organizations. In the first portion of the event, Borislav Popov, head of Semantic Annotation at Search at Ontotext (http://www.ontotext.com) will share: • Insights from a 5-year long experience of the BBC • Solutions for B2B publishers such as the Press Association • Semantic publishing platform for all large newspapers in the Netherlands • Internal content publishing at Siemens Atanas Kiryakov, founder and CEO of Ontotext (http://www.ontotext.com), will take you on a more technical journey into the working day of a semantic database as back-end for the BBC Sports website, handling millions of requests and thousands of updates, which makes it the highest profile use case for a triple-store today. Atanas will also cover industry efforts into performance benchmarks for RDF databases. After a short break, several companies will discuss their business challenges and why they have chosen semantic technology to solve such challenges. ____________________ First part of this session will be driven by a presentation of Borislav Popov, head of semantic annotation and search at Ontotext (http://www.ontotext.com). Leading the semantic publishing product development and solutions in the recent years, Borislav delivered to clients like BBC, Oxford University Press, and Euromoney. Passionate about text analytics, semantic annotation and search. Leading a team of over 20 excellent scientists and developers primarily focused on semantic publishing. The second presenter will be Atanas Kiryakov, founder and executive director of Ontotext (http://www.ontotext.com). Product manager of OWLIM until 2010, Atanas is leading expert in semantic databases, author of multiple publications and book chapters, such as "Semantic Search" in "Information Retrieval - Searching in the 21st Century" and "Storing the Semantic Web: Repositories" in the "Handbook of Semantic Web Technologies"
- The Working Ontologist Dean Allemang & MarkLogic's Stephen Buxton
A special event with Dean Allemang & Stephen Buxton SEMANTIC WEB for the WORKING ONTOLOGIST with Dean Allemang http://workingontologist.org/ Dean Allemang is one of the world's leading web ontologists and CEO / Principal Consultant at Working Ontologist, LLC, where he helps companies deploy Semantic Web solutions. Services include training, keynote talks, solution planning, technology selection, data conversion, ontology architecture and modeling. Dr. Allemang explores more about the Semantic Web on his blog. http://dallemang.typepad.com/ Semantics is the future of Search. And Search is the future of Semantics with Stephen Buxton The BBC introduced the notion of Dynamic Semantic Publishing with its Sports web site platform - an award-winning platform that served rich, dynamic, personalized web pages during the World Cup 2010, and then broke records with its reporting of the 2012 Olympics. At the heart of this platform are two databases - one for managing documents and statistics (MarkLogic), the other for managing facts and relationships. What if those two databases were one? What new applications - impossible or impractical today - could you build? MarkLogic is well known as a document-based NoSQL database, with an integrated search engine for very fast Big Data-scale search and facets. In this talk we'll look at what happens when you add Semantic Technologies - RDF and SPARQL - to MarkLogic. We'll explore some interesting combinations of traditional and semantic search, and we’ll talk about the rich applications that you can build when you have access to documents, values, and facts (triples (http://semanticweb.com/marklogic-7-vision-world-class-triple-store-and-world-beating-information-store_b37123)) in a single seamless platform. Stephen Buxton is Director of Product Management for Search and Semantics at MarkLogic, where he’s part of the team that builds the world’s only enterprise-grade NoSQL database . Stephen is co-author (with Jim Melton) of the authoritative book "Querying XML", and a contributor to "Database Design", a book in Morgan Kaufman's "Know It All" series. Before joining MarkLogic, Stephen was Director of Product Management for Text and XML at Oracle Corporation. http://www.marklogic.com/what-is-marklogic/ Short presentation John O'Donovan - CTO Financial Times (FT.com) Roberto García - rhizomik.net Notes: We would like to thank MarkLogic (http://www.marklogic.com/) for sponsoring Food and Drinks for this event as well. Registration is free if you would like to help with the organization of this event, just get in touch with Marco ([masked]).